Low-resolution Face Recognition in the Wild via Selective Knowledge Distillation

Abstract

Typically, the deployment of face recognition models in the wild needs to identify low-resolution faces with extremely low computational cost. To address this problem, a feasible solution is compressing a complex face model to achieve higher speed and lower memory at the cost of minimal performance drop. Inspired by that, this paper proposes a learning approach to recognize low-resolution faces via selective knowledge distillation. In this approach, a two-stream convolutional neural network (CNN) is first initialized to recognize high-resolution faces and resolution-degraded faces with a teacher stream and a student stream, respectively. The teacher stream is represented by a complex CNN for high-accuracy recognition, and the student stream is represented by a much simpler CNN for low-complexity recognition. To avoid significant performance drop at the student stream, we then selectively distil the most informative facial features from the teacher stream by solving a sparse graph optimization problem, which are then used to regularize the fine-tuning process of the student stream. In this way, the student stream is actually trained by simultaneously handling two tasks with limited computational resources: approximating the most informative facial cues via feature regression, and recovering the missing facial cues via low-resolution face classification. Experimental results show that the student stream performs impressively in recognizing low-resolution faces and costs only 0.15MB memory and runs at 418 faces per second on CPU and 9,433 faces per second on GPU.

References in corpus (11)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Distilling the Knowledge in a Neural Network
• FaceNet: A Unified Embedding for Face Recognition and Clustering
• TensorFlow: A system for large-scale machine learning
• FitNets: Hints for Thin Deep Nets
• Deep Learning Face Representation by Joint Identification-Verification
• DeepID3: Face Recognition with Very Deep Neural Networks
• On Low-Resolution Face Recognition in the Wild: Comparisons and New Techniques
• Range Loss for Deep Face Recognition with Long-tail
• Rocket Launching: A Universal and Efficient Framework for Training Well-performing Light Net
• Synthesis-based Robust Low Resolution Face Recognition

Cited by in corpus (20)

• Knowledge Distillation: A Survey
• Delving Deep into Label Smoothing
• Benchmarking Micro-action Recognition: Dataset, Methods, and Applications
• Backbones-Review: Feature Extraction Networks for Deep Learning and Deep Reinforcement Learning Approaches
• Cross-Resolution Learning for Face Recognition
• Hierarchical Deep CNN Feature Set-Based Representation Learning for Robust Cross-Resolution Face Recognition
• Cross-Quality LFW: A Database for Analyzing Cross-Resolution Image Face Recognition in Unconstrained Environments
• Graph-based Knowledge Distillation by Multi-head Attention Network
• Reasoning Graph Networks for Kinship Verification: from Star-shaped to Hierarchical
• Low-Resolution Object Recognition with Cross-Resolution Relational Contrastive Distillation
• Cross-Resolution Face Recognition via Prior-Aided Face Hallucination and Residual Knowledge Distillation
• Octuplet Loss: Make Face Recognition Robust to Image Resolution
• CATFace: Cross-Attribute-Guided Transformer with Self-Attention Distillation for Low-Quality Face Recognition
• Rethinking Knowledge Distillation in Collaborative Machine Learning: Memory, Knowledge, and Their Interactions
• Periocular Embedding Learning with Consistent Knowledge Distillation from Face
• 3D Human Pose, Shape and Texture from Low-Resolution Images and Videos
• 3D Human Shape and Pose from a Single Low-Resolution Image with Self-Supervised Learning
• Student Network Learning via Evolutionary Knowledge Distillation
• Dual Discriminator Adversarial Distillation for Data-free Model Compression
• Improving Route Choice Models by Incorporating Contextual Factors via Knowledge Distillation